RIP Workspaces – Ep. 225

Workspaces used to be a simple mental model: a shared place for semantic models, reports, and collaboration. Fabric is changing that fast—workspaces are now a home for many kinds of items, some created implicitly, and not all of them are meant to be ‘user-facing.’

In Ep. 225, Mike, Tommy, and Seth talk through what this shift means in practice: how to keep discovery sane, when to treat auto-created artifacts as internal plumbing, and which governance basics still matter most while the experience continues to evolve.

News & Announcements

• Power BI Theme Generator — New icons and theme uploads (and a great way to keep styling consistent as visuals gain more formatting surface area).
• Explicit Measures Podcast — Subscribe and browse the full episode backlog.
• Mike Carlo on LinkedIn — Connect with Mike for Power BI + Fabric updates.
• Seth Bauer on LinkedIn — Follow Seth for BI engineering and governance perspectives.
• Tommy Puglia on LinkedIn — Follow Tommy for the latest “from the field” platform findings.

Main Discussion

Topic: Workspaces in the Fabric era (and how to keep them usable)

Before they get into the bigger workspace conversation, the crew hits a few ‘what’s new’ items—like the new card visual and how tools like the Theme Generator help you standardize those capabilities across a portfolio. From there, the discussion pivots to the real pain point: as Fabric adds more experiences, the workspace is no longer just a Power BI container—it’s becoming the place where everything shows up.

That shift is powerful, but it changes the day-to-day: more objects, more automatic artifacts, and more chances for teams to get lost (or accidentally treat internal plumbing as something end users should click). The takeaway isn’t ‘wait until it settles’—it’s to adopt lightweight standards now so you can move fast without turning every workspace into a junk drawer.

Key takeaways:

• Assume workspaces will continue to accumulate new item types—build a naming + foldering convention that scales past ‘datasets and reports.’
• Decide which artifacts are ‘implementation details’ and should be hidden/ignored, versus ‘products’ that need ownership and documentation.
• Treat auto-created items as governance signals: if the platform creates it, your team needs a policy for who owns it and how it’s managed.
• Standardize visuals and formatting early (themes, templates, default settings) so new capabilities don’t create a styling free-for-all.
• Use idea backlogs and community templates as early-warning systems for what’s coming—but validate before you operationalize.
• Keep responsibilities explicit: who’s accountable for workspace structure, security groups, and lifecycle cleanup.
• When the UI changes, fall back on your operating model: ship value, review regularly, and keep the workspace navigable for humans.

Looking Forward

Pick one workspace ‘hygiene’ standard this week (naming + folders, ownership tags, or a cleanup cadence) and roll it out—because Fabric’s pace won’t slow down for your taxonomy.

Episode Transcript

0:29 good morning and welcome back to the explicit measures podcast with Tommy Seth and Mike good morning happy Tuesday gentlemen here we go it is another week the summer has officially started for those of you with children exactly what exactly the next level of chaos has instilled yeah man like there’s something about the consistency of the school year where buses come and pick up children on

1:03 where buses come and pick up children on a everyday Cadence summer just brings its own new possibilities of karting kids all over the place along with the normal normal workload you have been you have been officially designated the cruise director director what are we doing like every every morning what are we doing today where are we going how are you going to entertain me camps campson Sports Camps and days home when I was your age I used to go outside and play with sticks how

1:33 go outside and play with sticks how about you do that you want a bike go find one in the garbage garbage we’re that bad it wasn’t that rough we had bikes we were I was just walking the other day and when I was a kid or when when my father was a kid he was talking about what would they do when what they would find random things to go do and they would do a lot of things on bikes bike was like a main mode of kid transportation and now there’s like a million different modes of transfer transportation we see

2:04 modes of transfer transportation we see like my kids are out on scooters and we have bikes and we have skateboards and like all these things didn’t exist a long time ago literally all you did was if you wanted to go somewhere you got on a bike and you went there like there wasn’t like you couldn’t choose like a scooter like all these other movement devices I was like it’s interesting maybe maybe kids are used to getting shoveled around in cars so much now that we can have these other vehicles that are less efficient right a bike was efficient you could go miles on a bike yeah I don’t think I would want I

2:34 yeah I don’t think I would want I wouldn’t want to Pedal miles on a skateboard or like a scooter like that would be a mess or the Big Wheels man don’t follow the Power Wheels oh the big wheels the thing in the front yeah yeah those were fun yeah those were those were the best until you blew the tire in the back and then they were garbage right yeah yeah yeah that’s funny good times good times good times yeah Summer started for us a few weeks ago there’s pool and swimming that

3:04 weeks ago there’s pool and swimming that the girl the kids are doing now there’s a few camps basketball camps where do you go to where do you go to swim you have some friends that have pools or do you go to like a guy no we’re teaching them swimming so you’re going to swimming lessons yeah yeah yeah yeah yeah so I think that’s a really good skill to learn I when I remember being a kid I have a lot of fond memories of just being able to go swimming and swimming in I lived in Maryland at the time so we had the Chesapeake Bay was where we had to go swimming we hang out there I can get friends pools that was just a lot of fun

3:34 friends pools that was just a lot of fun I would spend hours at a pool I love swimming all right no one else does all right Grace food’s great kids love swimming swimming lessons swimming at Camp it’s a swimming it’s great all right let’s talk about data things sections that we have opinions about I will say I am loving the the new card Visual and seeing what everyone’s doing with it it’s that is a great addition great addition

4:05 addition Miguel Myers is the one who’s producing a building creating these new layers of visuals and you’re seeing a lot of new features being added this one does a lot of things you can run the edges you can change the corners you can stylize a whole bunch of things on this card and add images and colors and man it is really really exciting to see that one so one so we are actively adding that to the theme generator here shortly so that should be out here in the schema yet is it a

4:35 out here in the schema yet is it a schema schema super stoked because this just to me is like the precursor of all the stuff they’re working on right yes you think about it it’s like it it probably a lot of those properties or things are super easy to implement like oh oh you want rounded corners on the top and yeah I can show that and blah blah but all it takes is somebody like Miguel to go in and be like can’t we just do this this and this oh yeah yeah here and all of a sudden it’s just it opens up a whole slew of possibilities for for folks to

5:07 slew of possibilities for for folks to customize and build their own thing within a visual that just automatically does it for them so if this is I’m excited because if that’s the direction that there’s a lot of more a lot of more cool looking reports coming down the road the only thing I would add to this one is is with with the ability to make a very flexible card it becomes more challenging to know what you can possibly do with it there’s always this person who says I’d like to see the art of the possible and so there’s always someone who’s

5:38 and so there’s always someone who’s asking for like what could be done with this so you really have to for example in the card visually I can do some simple things you got to study up on it you got to see what people are actually building with it and learn how they’re doing it so you have that knowledge of like understanding what you can possibly build so I like the card I think it’s really cool I’ll see if I can snag the link here for the new card visual and then here’s the blog post for that you’ve already been doing that like for think about the creativity of even Miguel right like and in the blog lays

6:08 Miguel right like and in the blog lays out like this would be 10 different components that you would put together like we’ve been building like the what is possible with the rudimentary building blocks we have with visualization I think this just simplifies it maybe to your point now to a whole nother level where yes we can we can do a lot of different things well I had to do like I have to go into the card now and like understand all the settings in the card and this is why I think the theme generator makes a lot of sense because now you can just change the settings and see the card real-time updating in a theme file and

6:39 real-time updating in a theme file and then you can stylize it as you go so I go so I I really again I love the card I think it’s great really good function adding feature ability right and also being able to run one query all at once and return all the data in one big table you return all the data in one big table one if you get one row for each know one if you get one row for each metric you need which is great I think that makes a lot of sense almost every I would say probably about I’m gonna try and 75 to 90 of all the reports that I ever build has a probably more than one card on them

7:10 more than one card on them it’s it’s fairly common to have a series of kpis across the top of your Port either on the home page the overview page whatever you want to call it like there’s there’s usually more than one card I’m building I feel like anyways really like that visual any other things that stood out to you guys on the blog are recently around I don’t know if you guys saw but Alex Powers actually created a power query template to use in data flows to scrape

7:42 template to use in data flows to scrape the fabric ideas site I saw that one yeah and then I let me see if I can find the link but I think it was someone else who then basically is creating a fabric to create a data framed creating like a one like for the whole idea site if I could find it which I will find it all right well a tummy take around for the length of that one one the other things will bring you along for the ride for that yeah exactly

8:15 for the ride for that yeah exactly watch Tommy’s search internet Verizon another announcement from the blog was was the Forester announcement was augmented bi platforms Q2 2023 I was out there now so Microsoft just announced that they are again named another leader in the Forester wave which is again they’ve been a leader here for quite a while but I think it was interesting the comment or quotation they pulled out of the report themselves Seth you want to

8:45 the report themselves Seth you want to read the quotation because I think it was funny there you picked up on it I didn’t pick up on this oh yeah so Forrester noted in the report it’s close to impossible to poke holes in Microsoft’s overall bi strategy which includes Vision Innovation roadmap partner ecosystem adoption and pricing flexibility and transparency that’s all very true and this is one of the reasons why they’re pricing flexibility is the issue Segway last last final sentence gave me the Chuck over it isn’t that

9:16 gave me the Chuck over it isn’t that Microsoft is bringing wildly different capabilities to the market it’s that it has the vision and Deep Pockets to invest in all the capabilities to do them all quite well it’s like the Deep Pockets part was like it’s a good job of Tableau a jab almost yes it was but at the same time like it’s it’s not like Microsoft wrong mildly different capabilities I’m like well in certain certain areas are there tools and things that have been out there that have been serving a lot of the purposes that they’ve pulled together all

9:48 that they’ve pulled together all within Fabric and their ecosystem yes but just like power bi right it’s it’s it’s how you’ve brought these things together and how well they work together so the preview of fabric super exciting we obviously talk about it all the time I think I think I think this just accelerates really fast the way power bi did right as an ecosystem ecosystem introducing a new way in which businesses interact with data and I think we’re going to talk about that today right especially with how does

10:18 today right especially with how does that landscape change in in the modification of our understanding of like workspaces I wonder how long forrester’s been doing augmented business intelligence

10:33 augmented business intelligence platforms I know that’s not garnered it’s always been like a bi platform but the augmented side of that seems to be a little newer I don’t know I haven’t I don’t follow Forester very much other than the fact that they every so often pop out a little review I do think it’s very interesting so if you go to the article which I’ll put out here as well this is forester’s report actually on so so or restored friends about that right Foresters report yeah there’s actually a really interesting graphic that’s a halfway down it talks

11:03 graphic that’s a halfway down it talks through like here’s the weighted averaging it talks about Amazon Domo Google IBM Microsoft microstrategy Oracle and points out all of them the strengths of them and they very much like the strategy and the market presence of Microsoft but they basically gave them 100 scores all the way through the group there but it’s also interesting to see where does the other tools integrate with different Services their current offering is is where the price the price the scores really start to differ

11:34 scores really start to differ I don’t think I’ve ever seen like it the visual is interesting along with the table but like fives across the board like they basically pegged the chart oh yeah especially for that first graphic yeah but if you think about like even and that’s that’s probably why to me the messaging was a little little confusing around like what are we getting what what’s coming up and what’s longer term Vision Fabric and everything within it but it like this is this is more focused I think towards the

12:05 this is more focused I think towards the strategy and AI Innovation like how how are we going to accelerate the the traditional business intelligence or even the second wave that we’ve been riding with pulling in large data sources but still having to plug into multiple different services to do that so leveraging all this together I think is an interesting move to me especially around one Lake consolidating data like I in in my basic understanding

12:35 data like I in in my basic understanding of things I would just assume that makes it a whole lot easier to start to leverage the advanced capabilities that will start to have in in co-pilots and AI on that ecosystem so it’s interesting you mentioned co-pilot because I was looking at the Microsoft blog and I actually on the main article that’s on the main page May 23rd they still have the introducing Microsoft Fabric and co-pilot in Microsoft power bi’s blog and when I go into that blog there’s like literally a one paragraph

13:05 there’s like literally a one paragraph around co-pilot and everything else about fabric I’m like I feel like the title was a little bit misleading you’re focusing on co-pilot and you only have like one paragraph on what co-pilot’s gonna do gonna do so look yeah it’s a little bit like I feel like you’re you’re I feel like you’re trying to emphasize the AI side of things a bit much it’s it’s almost like naming a podcast show rip workspaces yeah workspaces Microsoft WTF yeah exactly workspaces

13:35 exactly workspaces the the fabric right exactly fabric the fabric yeah it’s going to change things in in the fabric so I think this is a good transition into our our main topic for today so let’s jump in so our topic for today is rest in peace RBI workspaces so this will be an interesting topic to see where people land on this one I have some opinions around this one and we’ll see what happens here I think the addition of Microsoft fabric

14:05 think the addition of Microsoft fabric has is going to change a lot of patterns for organizations on how they want to think about their data and how they’re going to be able to engineer information so maybe we should start at the very slightly at the beginning here let’s talk about workspaces and what change our workspace is going through now with the addition of fabric what have we seen feature wise come out here recently just give me a quick overview of why is it why does a workspace exist the biggest things with workspaces since

14:36 the biggest things with workspaces since the beginning was always for one one drive teams and groups to be great oh no never mind you’re going way back yeah so really workspaces have always been part of power bi I’ve honestly since they launched it and really it was a way to simply from a collaboration point of view share manage and organize content be this was before power bi apps there was a thing called content packs back in the day too it was rarely meant

15:08 back in the day too it was rarely meant to for if I’m a particular user I can publish a workspace like I say a sales workspace I have that ability and I then I can share within that workspace individual items they’ve always had an idea of roles and I think it was back in the day it was literally just admin and member but it’s really just the way to organize content that way in a larger organization people who don’t need to publish or manage or edit or delete content in a workspace that they are not really responsible for they don’t have

15:39 really responsible for they don’t have to worry now workspace is really have grown obviously evolved to being tied to Power by apps we will not talk about content packs at all but they’ve deployment pipelines really being part of the life cycle of publishing deploying content and then also sharing and letting others have access to that content there’s still the idea of Administrators and contributors to create publish and manage the content

16:09 create publish and manage the content but now there’s the other side of beams access and see what content is available in a workspace so Oklahoma no you you mixed a lot of things in there and I’m wondering did you like with workspaces did you ever share Organics workspace access with just a viewer somebody who’s never okay hold on I like this the distinction I want to make because you threw in share

16:40 want to make because you threw in share yes into this ecosystem and so where I was going to go workspaces are for the doers they’re the builders you’re you may be in a team where you’re collaborating with these objects but ultimately workspace access on most every level is specifically for the teams of Builders whether that’s in the business or in Enterprise or wherever it’s you or in Enterprise or wherever it’s where you’re creating these logical know where you’re creating these logical places that may be specific to

17:11 places that may be specific to a business area might be a test and a prod prod Etc but it you have access to the workspaces where you’re a part of managing or should have visibility into the products themselves to fix them they’re not the area where we bring in every consumer of a report that’s why we have sharing mechanisms directly from a report or with apps and now that those are extended they’re even better with the multiple different audiences

17:42 the multiple different audiences but it’s still the developer right like this is the the doers of building things like that’s who belongs in the workspace and the permissions that go along with them them in the Microsoft ecosystem for I think we’re talking specifically power bi right yes yes so here here’s where I want to start talking about like man there’s so many things that go along with this I feel like but Tommy you brought up this whole sharing portion I think before apps powerbi. com was a good

18:14 think before apps powerbi. com was a good method to share content through I think that was like the original yeah thought here but things got more complicated when we started talking about okay how do we how do we start leveraging corporate or certified data sets and reports with in combination with like non-certified things and again I’m thinking through that Matthew roaches pyramid again right we have at the top we have Enterprise data where data warehousing or Enterprise data sets and then we have down at the very bottom we have like personal or Team level

18:44 have like personal or Team level reporting right so on one end of the spectrum we have a very controlled experience where no one should be touching things in different environments like you it’s literally a view only ecosystem and then on the other end of the spectrum we have this you have full control build whatever you want it’s it’s on you and so as I look across workspaces I think about workspaces as we and a pattern that emerged from Microsoft Microsoft and even on the adoption roadmap was this pattern of have a single workspace

19:14 this pattern of have a single workspace or a deployment pipeline for just the data sets and make a separate deployment pipeline a separate group of workspaces for the reports or the thin reports so now you have a data set that can be migrated between Dev test and production and then you have a thin report that can be going from Dev test and production I think this makes a lot of sense when we’re talking about Enterprise bi you we’re talking about Enterprise bi the top that pyramid very governed know the top that pyramid very governed managed and secured I’m not sure I always agree with this at

19:44 I’m not sure I always agree with this at the lower end of the spectrum and so this is where I’m not struggling but I’m continually teasing out workspaces to me feel more like a pairing with the app so when I share or distribute content I’m more likely to not get access to power bi workspaces and instead give more access to the app level stuff so I will publish an app off of the test workspace and have people go test or quality check your data there and then on the production environment

20:15 and then on the production environment again we don’t share the actual reports directly you actually keep that workspace locked down to maybe one or two people the release manager right that Persona the person who’s like moving content through these different pipelines and and the app is the only way that you’re sharing information and I think to your point Seth I think the to me the workspaces it should be really focused on the people doing the work executing the movement of content and making sure that content is correct and out the door yeah and the

20:45 correct and out the door yeah and the thing I like about the way they implemented apps even in in the fact when it was one workspace one app which was less useful yes now now I like with multiple audiences and different artifacts from a workspace I think it it meets all my needs but one of the things I like is I can deploy a change to that workspace and it

21:05 deploy a change to that workspace and it doesn’t automatically get rolled into the app right I guess go update the app before the changes that I’ve deployed into the service from the Report perspective are actually pushed in to the an end user and that that almost acts like a uat right push do you find checks and validation right before you’re gonna light it up for them so here here’s where I want to start poking on things a little bit more around this so okay we have this General concept around here’s an Enterprise data set where do you where

21:36 Enterprise data set where do you where do you feel this fits with like that team or that personal level workspace so so here Here Comes part of my like I I understand that more of that stuff at the lower end of the ungoverned space of content it’s just going to happen happen so in your organizations or things that you’ve worked with with clients how do you feel it’s better to produce or how do you where does the workspace fit in those more

22:06 those more personalized and and and Team level reporting what does that look like to you and does and does that Team level reporting or workspace does that hang off of data sets that have been built and developed by Enterprise in current state power bi in current state power bi where do you feel like that fits or do you think it or maybe it doesn’t fit I don’t know did I catch you off guard no you’re asking me specifically not us where I’m just like waiting for Tommy to

22:37 where I’m just like waiting for Tommy to to chat I’m Gonna Let You Go on this results are you not with us yet okay I am with you so like the the use cases I think there’s many different use cases and one one that you outline is not one that I I typically am I personally am implementing because we’re my team is very Enterprise focused right yeah yeah so you’re building everything at the top pretty much correct it all comes through one team and it’s gonna start but I feel like that’s how it starts I feel like it starts with that all right you have a single handful of data sets you have a

23:08 single handful of data sets you have a handful of reports that start Enterprise and then you’re gonna have this request well the request will come I think right I think there’s another wave of requests that are going to come hey how can I get access to that data how can I build off of that data set or can I take a copy of that report and build my own thing are you gonna say no to all that this is a conversation we’ve had way too many times from the cell service to how much access and it’s tough especially because there’s no real automated

23:38 because there’s no real automated automated way especially for an Enterprise organizations to manage this if you’re a small business or a smaller company Maybe 30 40 workspaces yeah it’s probably easy to have maybe like a request system but if you’re a larger larger organization it’s really hard to say okay who is who can administer a workspace yes who’s responsible for giving those okay I like this point and and then because you think about the also the levels of the the

24:10 the also the levels of the the Champions where you have obviously the Federated bi team but you’re gonna have people who are self-service or do they get to publish to those same workspaces I like this question a lot so okay so like can you go ahead point to clarification we’re talking about non-fabric workspaces right now just power bi or all of a sudden are we mixing in capabilities that fabric allows us to do okay so right now I

24:40 allows us to do okay so right now I really want to make a distinction yeah yeah we haven’t yet for any listener is like what are the differences of evolution of your just in power bi how we operate with workspaces what are the other things that are now implemented in fabric which opens up a whole slew of conversations which I’m fine digging into but I’m I’m having a hard time tracking whether or not we’re jumping right into that makes little sense so if we’re still in power bi

25:10 sense so if we’re still in power bi realm without no no up until this point I’ve been thinking solely around this I haven’t even touched what fabric changes to this ecosystem I’m just thinking about okay ignoring all these things that come with fabric a lake house data engineering data science notebooks pipeline all this other stuff just gets added into into this whole ecosystem and so we’re just talking about the complications already of workspaces with him yeah I’m just I’m just thinking about I think a lot of so where I see things going I think a lot of organizations don’t have a fundamental grasp of what is a workspace admin what

25:42 grasp of what is a workspace admin what is a workspace member contributor and viewer and I would argue the the admin of a workspace is the person who would request it for their team or their individual group and they by requesting a workspace are default the admin and the owner they have they need to go through additional training they need to understand what it means to make an app they need to understand deploying things they need to they need to own that content so when we talk about self-service bi and this is and this is where I think I’m starting go with some

26:12 where I think I’m starting go with some of this these thoughts or this Theory here right so Seth you’re in a space where it’s a totally Enterprise right you own it your team has admin privileges and all these things and no one else has basically contributor or viewer access for any of your environments because you are owning that content at some point there will be other teams that need to own a workspace and they will have to figure we have to like someone’s gotta we gotta like go when there’s a problem with that workspace and the data comes out incorrectly who do we go to to say hey you’re responsible for the stuff coming

26:42 you’re responsible for the stuff coming out of this workspace what does that look like and so that’s where I think like the admin and the member portions of this make more sense and I would say 90 of all your people should always be just a contributor and I don’t ever see a use case really for the viewer role anymore now that I want to use apps to publish content so that’s where I was going with some of that that talk around who can administer the workspace what is the responsibility and I feel like the workspace for me has

27:13 I feel like the workspace for me has become more of a delegation of responsibility right so if I have a data set that’s sitting in a pipeline that data set is owned by someone the owner of that data center the person that I go contact when there’s a problem with that one is the admin or potentially a member of that workspace that’s where the pro that’s where the the group owns that stuff stuff if I then give access to that data set to another business team let’s say I’m in sales right and I have the sales data set that’s made at the corporate level and I start giving the sales team their

27:43 and I start giving the sales team their own workspace and they can then connect to my data set that’s corporate I think that’s something that Microsoft wants you to do they want you to design these team based workspaces of people building content on top of models that are Enterprise grade and to me this is where I think the challenge comes is we now have to as a as an admin or an owner of the power bi workspace we now need to have this clear handoff this this moment in time that says hey we don’t own your content you are the

28:13 we don’t own your content you are the owner of this workspace and here’s the expectations that goes along with it owning this workspace because I feel like too far too often more organizations just give out workspaces and there’s no Rhyme or Reason there’s no thought to it there’s no commitment of ownership to these things and so then when data goes wrong or data has incorrect there’s no support mechanism to support these things longer term term I like the point because the way I would look at that is you may not necessarily need to like

28:43 you may not necessarily need to like tell people that they can’t build a workspace however yes if it seems prudent to have a process by which people would know how to move their workspace and bingo managed work yes by admins yeah if you’re starting to leverage data sets or need access to certain things yes you automatically get put in a bucket where you’re part of a part of a capacity right how do I get my workspace into a capacity now you’re managed who’s who’s the owner of that is an admin

29:15 the owner of that is an admin automatically need to get added to the workspace or we just you managing your own refreshes Etc so I think I agree it it is it it is it’s a really good idea to either control those mechanisms or have a process that defines them for people building things yes and in general before extending into fabric it you should get much more serious around the administration of what’s going on in the world yes as far as decimals are concerned okay yeah I

29:46 as decimals are concerned okay yeah I like that a lot yeah because I think I think if you have a loose management of your control of administration and delegation of workspaces to other teams adding fabric is just going to add more fuel to that fire and without that proper at least in these at least a thought process around like how we’re going to govern these things and what does that mean because now we’ve added another whole layer of just complication

30:16 but here’s the thing though and the cell probably goes into the new domain feature too the admin it’s really hard to just have one admin obviously power bi and more or less an admin per workspace it’s it’s something that needs Automation and I think this is something that we haven’t talked about as much but the need for I this is really when you need it or those who are in charge of active directory to be involved because there’s no way this can be managed from an individual scale what do you what do

30:47 an individual scale what do you what do you mean by that so you said it’s really hard to make an admin for a workspace I’m not sure I understand your statement there there it’s not necessarily hard to make an admin for a workspace but it is hard to specify particularly one administrator for a workspace but having all the rules why because well think about this they’re they’re going to be what are they responsible for if they’re an administrator for our works okay this but that question is what an organization has to decide on what does

31:18 organization has to decide on what does that mean right I would I would argue an admin is responsible for controlling who has access to that workspace how content is being shared how data sets are being built and the quality of those data sets and the impact if that if that admin is on a premium workspace that admin is also

31:37 premium workspace that admin is also responsible for is your data set eating up all of our premium capacity and we will have monitoring that will let us know when that workspace becomes abusive and someone loads a mega big data set or a mega bad query like so the admin is the person we go to to start that conversation to figure out hey there’s three data sets in this in this workspace that are causing us all kinds of headache and it’s causing our other queries to time out because your query or your data set is so bad it’s eating up all our premium capacity so there has

32:07 up all our premium capacity so there has to be one or two people at the top of that workspace that says you are the contact person but I think this is this is something that we’re not communicating to owners or admins of workspaces and when I show up to companies it’s tons of admins there’s like six admins seven admins like everyone’s an admin and no one’s a member or a contributor well who do I talk to talk to can I make a recommendation okay let’s let’s parking lot this way do a show on roles of individuals and

32:38 do a show on roles of individuals and workspaces yeah yeah I agree with that one separate thing than we want okay today so we’re high level I get you I’m with you on that one so taking that aside so I think there’s things to tease out here the reason I wanted to bring this whole data set and workspace and pipelines and all this stuff is now you add in the context of fabric and so now I feel like there’s another whole layer of complication that’s occurring here where we’re saying okay in previous designs I’d have a pipeline for the data

33:08 designs I’d have a pipeline for the data set I’d have a pipeline for the reports do I now need a pipeline for the data Lake and the data engineering that’s occurring behind all of this so instead of talking about six workspaces we’re now talking about nine of them of them holy smokes no no like okay okay so it’s it’s right it’s a prospective use case right okay I think I think the solid argument you made was start from a solid stance of understanding roles governance understanding of what workspaces are and

33:39 understanding of what workspaces are and do in your environment before you open that fabric trigger yes I agree with that 100 and I think that’s it deserves more like discussion because the Nuance of roles and responsibilities I think it changes changes within the context of like what your capabilities workspaces are driving it in a little higher level what like we’ve defined workspaces within the power bi ecosystem when you turn on the fabric when you turn on the fact when you put

34:09 when you turn on the fact when you put the fabric on it feels good what are what are some of the new capabilities or things that are ancillary around workspaces that are now introduced into that ecosystem that people need to be aware of aware of so I think the first one Tommy and this may not be a feature of this is domains so there’s I don’t know if this is a fabric only feature it feels like this is a power bi feature that was added at the same time as fabric so it feels like a fabric fabric feature but I don’t think this is only linked to fabric really I think so I think you can use

34:40 really I think so I think you can use domains I think you can use a domain it’s a domain to find a domain there’s a new feature in the admin portal that when or even when you go to settings on a workspace you can now select a workspace and then you can pick from a list of domains if you think of domain being a collection of workspaces that all relate to the same team or subject matter within your organization I may have a sales team and there may be different components of that sales team and we may have five workspaces of team members that are going to build things in power bi and so I see this is how I

35:12 in power bi and so I see this is how I interpret it domains are a collection of workspaces that all pertain to sales that’s how I would see that correct and so they allow you to easily group or a particular use case right yes you had multiple different workspaces but they’re all controlled by sales this would allow you to control that that particular area I fact check on the I think I thought I thought domains were fabric only I think it is because it is the administrator role

35:44 because it is the administrator role but everything at least from the documentation says is called fabric domains domains not necessarily just single domains and this is where it gets confusing to me because this is fat because like it doesn’t if you go to the admin page it mentions nothing about fabric fabric and you can create a domain and you can add non-fabric workspaces into a domain so I think it’s just a logical grouper but they’re just calling it fabric because now power bi is Fabric or the

36:16 because now power bi is Fabric or the website of power bi is now fabric I don’t really I don’t know where the lines draw at anymore at this point like it’s going to be like this either way there’s this concept called domains and you can associate workspaces to them yes the the interesting hierarchy to me here is is this is a logical grouping and you also have a logical grouping of workspaces but if you think about it you could you could associate multiple different capacities two different workspaces right so it

36:46 two different workspaces right so it depends on what the needs are or whatever how how those get billed or charged or whatever yeah totally those all still fall under the domain of sales so sales needs multiple capacities or they need to do certain things because well well whatever pick your business unit right like but this makes sense though because if you want to throw like a premium capacity on like if if sales has bought their own premium SKU and they’re going to pay for it

37:16 to pay for it all the workspaces you would you would want to manage all the workspaces that just live in that premium capacity as a collective group the settings would need to be on the same it could be a mess if you had five workspaces all attached to the same P1 capacity and you have to go through and change okay we’re gonna have ai off we’re gonna have data flows on we’re gonna have this look there’s all these settings that would make sense and I think that logically makes sense the idea is the domain is you can apply group settings or settings to the entire library of workspaces that make up that

37:46 library of workspaces that make up that domain and that’s the weird part here because it really is just in a sense it’s not it’s really less about the organization where you think maybe it’s like a hierarchy of folders it it is that but from a UI from a collaborator point of view it really has not a lot to do with that’s really from a high level security point of view at least from the data Hub right where I can’t see the necessarily the domain in a workspace

38:17 necessarily the domain in a workspace obviously first I need even if I have access to a domain I still need access to the workspace well yeah and that’s like but I think that’s part of the role conversation too right like how do you how do you segment out there are different levels of permissioning within domains as well right and there are different different criteria so once again like who are the people what are the roles and responsibilities Etc I think for the context of this conversation it’s a bucket and we can add a bunch of workspaces onto it and and that allows us to extend the features further other

38:48 us to extend the features further other things that are different with fabric workspaces than power bi right is you have access to all of these other features within the context of a workspace right which I think is the big change for report authors and users is that we’re now opening up like all of the ETL and data and sources of information to people in that developer workspace and that’s where I think

39:19 and that’s where I think a lot of the like oh my goodness reactions come because it’s a lot you’re introducing the potential of more individuals to be added to a workspace instead of just the report group and and domains are also interesting because a domain now also has an admin and a contributor role so we’ve added another layer to like what does that mean mean how does that fit with everything else

39:51 how does that fit with everything else I just my general feel here is and the reason why I I was actually pushing for a lot of the rip workspaces I feel like I’m saying rip workspaces we are now going to be inundated with workspaces like it’s going to be workspaces for everything there’s going to be like yeah we had challenges before but now this adds another whole layer and this is why I think the domains thing appeared the feature applied because now we have a ton of permutations of what we can build with domains and and so now adding

40:22 domains and and so now adding when we’re unable to get folders before but now we have this domains folder a capability potentially right yeah which is also like why I want to walk away from getting stuck into roles specifically sure because those are those are the people right like in focus more on more on the the big shift between what workspaces mean right and yes workspaces in power bi are like Tommy said earlier that they’re the

40:52 like Tommy said earlier that they’re the contributors the administrators the members of reports right the final product that’s going out to the business we’re we’re fabric oh I see what you’re going with us I like this we’re fabric challenges that is what is what is the problem or challenge with power bi on its own in in workspaces and contributors it it’s the final product it’s it’s a data Mart or it’s it’s a model that somebody else

41:24 or it’s it’s a model that somebody else can connect to and leverage and use but we know that if I’m sales and just because I give them access to a fully curated model they’re going to have a side thing that they need to connect to that they’re not gonna have like there’s always another bit that some other part of the business requires on the data set correct so so where where I think this evolution of power bi being this ecosystem and now fabric opening up

41:55 fabric opening up all the layers of data yes to the same people I think challenges or introduces a completely different View that businesses need to like

42:09 that businesses need to like digest and figure out how they reorganize or if they do the like the like if they do of who gets access to the workspace and how do we analyze and verify data because ultimately what this should ultimately fabric is from from my perception of the marketing everything bringing everything to a unified place and that truly is the case if I think about well what is the challenge of power bi data sets and workspace and

42:40 power bi data sets and workspace and contributors right now is It’s the final It’s the final product always but if I now open up the layers of like okay here’s where I have the data you don’t need to care how I did it but if you want to walk through how I did it this is like now I have this interface that would allow me to walk somebody to the objects that if they need like to go access before a data model but it’s still part of a certified layer like now I have that easily available to

43:10 like now I have that easily available to me in the one Lake and I can share that with somebody if somebody needs to vet business logic in ETL we can walk through that like you’re all within this interface where now it’s just hey rather than focusing so much on a final product when data people can work off that but many times require the other thing or I need to hook into the like a back-end system those are challenges that we’ve talked about and

43:41 challenges that we’ve talked about and faced in like how do we curate something that’s like ready for the business or certified to a certain level and how do I start leveraging that how do they see like the Enterprise cleansing of a whole bunch of data and getting to a golden record those were all separations of I. T and business and where this gets interesting to me in this workspace level is are we flattening that out and actually like can we walk through use cases where

44:11 like can we walk through use cases where this now opens up a whole new realm where I can potentially bring people I would have granted access to these data sets to give them access to more data via one Lake Etc and I wrote down it’s enough questions or I wrote it on a ton of comments go ahead Tommy you go first no and I think the biggest thing with the the domains is very again I at least from what I’m seeing is the focus from the accessibility or discoverability not so

44:43 accessibility or discoverability not so much the security that you would do with workspace admins and the contributor roles because there’s no really role in domains except the administrator who can assign a workspace to a domain so I to me is really the what they’re trying to do with the domains it has nothing to do with from the collaboration point of view I don’t think anything changes but because if you go to the like the one like data Hub previously the GitHub in the power bi service created with me that’s where

45:14 service created with me that’s where you’re gonna be able to see not just your data sets but you’d be able to see anything anything any content within that domain so rather than a user trying to hey let me see the this workspace or this workspace it’s just the marketing domain wait a minute wait a minute so you’re saying let me just heard you say is it are you saying that if I’m in if I’m a contributor to a domain and there’s another workspace I don’t have access to Do I by default get to see the data

45:44 Do I by default get to see the data created in that workspace if you’re absolutely did I hear you wrong or is that what you is that what you said or no so with the domains or is it like cascading requirements like you’d have to be and contributor to The Domain to see everything in the domain and a contributor on the workspace to see the content created in that domain well you still have to assign workspaces to that domain that’s actually a really good

46:14 that’s actually a really good question Mike just the way you were describing it to me made sense it makes sense maybe maybe that makes sense right there’s a cascading effect of permissions that would need to occur between a domain and multiple workspaces I’m not sure yet sorry I didn’t mean to derail your thoughts no no that’s a great there’s a great question so find items for the workspace you don’t have to figure it out right now no no I don’t think yeah regardless it’s for the again I think

46:44 regardless it’s for the again I think it’s like a more of a UI thing because I don’t think there’s anything in the admin portal or for like all since you give a you create a domain and then now everyone has access but again I haven’t tested it out yeah I have to go I’ll build something like that where I have you something like that where I have three workspaces in a domain and know three workspaces in a domain and then what happens if I give contributor access to the domain level what can you see but to your point though Tommy I think what you were describing there was making a lot of sense to me and Seth I think I want to pick on your points here a lot as well right there’s new the

47:14 a lot as well right there’s new the fabric ecosystem is now bringing two new personas into the power bi ecosystem we have not had before typically we’ve had the model builder and the report Builder there’s been two main things that we control at the power biot. com level now what we’ve basically bolted on here is azure data Factory synapse and SQL Server they’ve all been brought directly into the ecosystem now power bi and now we really do have what I would call A A true data engineering role that can live

47:44 true data engineering role that can live inside the power bi ecosystem and so so the so the new workspaces this lake house item and I was thinking this myself right we now before we had model level access permissions and Report level access permissions and potentially we would manage those with workspaces but now what you’re describing set to me sounds like there’s another whole layer of access here we can go we can now provide access to individuals all the way down to the table level and so when we’re talking about

48:14 talking about what we can do in data flow is what we can now do with notebooks we can access those tables provide a link or a shortcut to the source of data and then build Downstream engineering so to your point Seth right you said Hey look the the sales team or whatever that team is working on their data set you can give them a data model but they’re always going to want to add some of their own data right some of that stuff is going to be certified some of it’s not going to be certified so now we can certify individual tables and say hey

48:44 certify individual tables and say hey marketing team here’s a model that’s certified off these certified tables here we go although although if you need additional access Mr power bi Champion here you go here’s the table that we built the model from you can go access the raw table and add or append any data you want and again I would argue back is this is another use case where you need to think about what does the administration of that table look like and how does that play out inside your ecosystem are you going to let people

49:15 ecosystem are you going to let people build their own engineering after the certified table and then what happens to those data sets and reports Downstream of that does that mean those things do not become certified and that’s not Enterprise grade or is it just for that team so you you now have the ability of having more of that access to raw information and what I see Power bi doing is we’re marching the business user further and further and further up the stream closer to where the source of data lives which I think is good but I also think

49:46 which I think is good but I also think it’s without thought on how you’re going to govern it or think about who is responsible the data stewardship conversation becomes much more important now yeah that’s where that’s where I need to get stuck into if we’re if we allow them access to some some of the certified tables right correct say rather than a certified data set and that’s all you can interact with now it’s a hey all devices for our certified data sets come from these tables you can have access to them and bolt-on whatever

50:17 have access to them and bolt-on whatever you need to about the value of direct like hypothetically would be well now I don’t need to worry about everybody importing the same thing over and over exactly it still comes from it’s a shortcut and if they add additional artifacts artifacts into their report that would be interesting to me to see like can can an admin see that like if we’re ripping through and we’re making sure that our our business is operating at Peak efficiency capacities not being like

50:47 efficiency capacities not being like absolutely killed or used or potentially is sales introducing something that should be in my certified data set and now I have a task for the Engineering Group to go make sure that it is and then alter that report right away yeah like those are the like the pieces of how how and where things are whether that’s going to be blocked by purview or whether or not we can see that lineage I think would be one of the strongest things things within this ecosystem to see for sure but either way I think it challenges

51:18 but either way I think it challenges not challenges but opens the door for possibilities of solving a lot of the problems that we would have that related to like you you can only access this thing in the reporting layer instead of backwards One Step where you would want to actually or maybe you only need a subset of the information but it’s still based on our something that has gone through a lot of rigor to cleanse and clean the data to the point that makes sense for a larger

51:49 the point that makes sense for a larger swath of business yes and I think the other thing which we’ll have to flesh out I think as a side probably in that roles conversation is with with these certified data sets right like are are should we be utilizing like the same admin group like group names that of workspaces when we permission things right and I wonder how far we could extend that back and forth because everything we’ve talked about is

52:19 because everything we’ve talked about is administrating via groups or group name in power bi because you don’t want to you don’t want to singularly manage users in that platform and that make that that like that reason times 100 now in fabric because it should be owned completely by azure 80 if at all

52:41 completely by azure 80 if at all possible because that’s how the entire ecosystem of security is going to be based and that shouldn’t be like it like when I say shouldn’t it would be a nightmare if you started adding single users everywhere oh yes try to unwind or manage that but it’s already happening I manage that but it’s already happening that’s the problem I think it’s mean that’s the problem I think it’s already out there people are adding individual users and they’re not using things like that for best practice and this is why I think not to go back to it again but like going back to the idea of a workspace and the admin yeah an admin can be individually added to a workspace

53:13 can be individually added to a workspace but I think that admin would then need to be responsible so with the responsibility of that admin portion that admin shows up oh by the way now you’re also admin of the members security group you’re also admin of the contributor Security Group you’re also admin of the viewer contributors viewer group for that workspace so potentially that admin now has the ability to add or remove individuals from the security groups that are now applied to that workspace so maybe that’s a process that you set up in place there that enables

53:45 you set up in place there that enables that admin person to be able to manage and push in or out those people which that also by using those security groups immediately provide access to any it support that you would may need to adjust or modify or adjust those those workspaces or those user groups so now you add more capability to your support team team so many thoughts here I can’t believe we’ve talked about this for 50 minutes already that’s why I want the table rules because man that’s gonna be I think this is just the first of

54:16 be I think this is just the first of many conversations but what I think the the value of of the even the name right of of our of our view of workspaces is when you when you log into Fabric or you open that fabric ecosystem you need to get rid of what workspaces were in in your head because the potentials grow exponentially depending on how you utilize them and what’s really intriguing and fun for me to think about and hopefully I’ll have

54:46 to think about and hopefully I’ll have some harder conversation points in the future is this opens the door to end-to-end Solutions within business units yes yeah and really it I think could potentially even challenge how organizations are structured because there could be a lot more of Engineers having access to multiple different workspaces that work in coincide coincide with these people but at the same time it’s not just locked into a power bi

55:17 into a power bi file all the time or power query like now I hey you need us to hop in and help you guys out let’s do it in your workspace you own it right but we’ll build this Automation and this this ecosystem that everybody is just leveraging and using and if it becomes big enough maybe they have to fill that own role for themselves right and they have their own person that manages all their ETL but across the board maybe that is it removes the barrier I think between I. T and business

55:49 barrier I think between I. T and business and really just it like leans into creating valuable data sets that are that can be leveraged by business users and if it means pulling them back one more layer where they have access to you more layer where they have access to data in one Lake and they just see know data in one Lake and they just see it as a table yep that’s that’s a win I I feel like what you just said there was it’s sounding to me like there’s a replacement of the SQL Server being removed we’re adding we’re adding to the ecosystem the concept of a one Lake and

56:20 ecosystem the concept of a one Lake and I feel like what’s what’s happening physically like like in the in the organization of companies is we’re now encouraging the business to hire a more robust data engineer in their team team because if you if a business unit has the need and can support the salary of that data engineer you’re right Seth you can you can build a a full solution in that business unit and they can maintain the whole thing all they real all the right well I I

56:51 all they real all the right well I I I you must be on a mission to mean I I you must be on a mission to replace SQL but I don’t think we’re not trying to replace it I don’t think we’re replacing the to me it’s not it this opens the door of it’s not how like what tools or how we create data sets I think it’s making these objects of value meaning these these data sets that are curated available to a wider audience and and that is the value

57:22 like creating those for the organization to be clear I’m not trying to talk about killing the language of SQL right I love it I think it works really well I just think that underlining technology that’s executing the SQL is substantially changing and and the idea here is like we’re moving away from the concept of like I need a SQL Server to do my work and that that was a very much dedicated I. T role that was there so the server is we’re less of emphasis on that and we’re more emphasis on Lake and connecting to things inside the lake elements yeah and it’s even more

57:52 lake elements yeah and it’s even more less about the services and the tools that you’re using to agree modify data it’s about data and I think that’s that’s what’s really intriguing to me about this conversation and workspace is and opening up doors to that is as we continue to talk about this those use cases all revolve around I think more options for the business to Leverage The outputs outputs the way they need to in these ecosystems and it becomes to your point less about

58:23 and it becomes to your point less about how we store it what tools we’re using because those decision points underneath are designed to scale and they don’t need to care about them yes I agree oh man this is good we are in a very interesting time right now and things have just and I think this is Alex Powers was on the last episode he was saying I feel like there’s been so much everyone’s trying to struggle to figure out where the bounds of this new world lives how does this new thing get governed or administered good thing it’s in preview

58:54 administered good thing it’s in preview for now it’s not officially released yet but like it we have time to figure out like what does this mean for organizations and how this how this will structure things all right so what I did is we’ll transition over to what does Bing chat say so I asked a very simple question over to Bing chat and I said should I use power bi workspaces to manage content in power bi I think chat GPT is a consultant the first thing it starts the first thing it starts with it goes it depends

59:29 so Bing you have to you put a you need to put a dollar in the kitty because you can’t start anything with it depends but it says it depends on your on your needs and preferences but Power bi workspaces can have some advantages according to the search results workspaces are used to collaborate with colleagues create a collection of dashboards reports and data sets you can also use workspaces to create and distribute apps to a broader audience some of the benefits of the workspace are granular workspace controls you have a contact list of people who have notifications about activities and

60:00 notifications about activities and issues but that’s a good point you can configure once a workspace OneDrive to use Microsoft 365 group interesting I would not have picked on that one but apparently Bing thinks it’s a good feature and you can share data sets between workspaces we talked about that one and you can create and they called it template apps that you can distribute to customers outside your organization which is true you could make a template app and and publish it on the app Source location it was interesting so I very good answer I thought pretty pretty on point there

60:31 I thought pretty pretty on point there Microsoft Microsoft on point on point I’ll give that a six I’d get five five out of what 30. what would you give it Seth any any rankings there oh yeah solid 7. 3 out of 10. 10. beautiful beautiful oh I’ll give it a probably like a 725 out of 1000. okay it’s good enough for me [Laughter] anyways with that we appreciate your

61:02 anyways with that we appreciate your listenership everyone thank you so much for hanging on while we blister about user roles and workspaces to be perfectly clear you’ve listened this entire episode I don’t think workspaces are dead they’re not they’re not gone I do think the controls and how we think about them needs to change so I do think there’s there’s an app there’s an area of Administration how you think about workspaces in current day yes should it should go by the way to your Naval fabric I agree I

61:33 the way to your Naval fabric I agree I think I think there’s going to be a a shift in mindset on how you leverage and use workspaces moving forward and it will you’ll probably need to adapt I would also argue if you’re are an admin of anything workspace related start thinking about these questions how do you want workspaces to be used and leveraged inside the context organization because these things I think should be at least defined up front and your team needs to be educated on what best practices and patterns that you feel like your organization needs to use because if you don’t do that you’re going to have always workspaces with a

62:04 going to have always workspaces with a whole bunch of admins and now you have a really hard time with data stewardship so I think there’s another layer of of data stewardship that’s going to be coming along with this as well kick the tires now in the preview figure it out yep test it out no all right that with that we only request that if you love this podcast please share it with somebody else let someone else you you found no value from this and you should never listen to it and it’s not fun and we don’t make jokes so Tommy where else can you find the podcast well you can find the podcast anywhere podcasts are available Apple Spotify

62:34 podcasts are available Apple Spotify make sure to subscribe go to Power bi tips Power bi dot tip slash podcast and you can submit a mail back what do you want us to talk about make sure to join us live every Tuesday and Thursday 7 30 a. m Central awesome thank you all so much and we’ll see you next time

63:12 thank you foreign

Thank You

Thanks for listening. If you enjoyed this episode, subscribe to the podcast and check out PowerBI.tips for more templates, themes, and Fabric guidance.